The overall goals of this proposal are to investigate the mechanisms of human hemoglobin switching, delineate the function of HSs of the LCR and develop a new approach for the study of the molecular and cellular control of human hemoglobin switching. In SA1, we will test the hypothesis that the human globin gene switching is controlled by a developmental clock-type of mechanism operating through a series of time-related chromatin modifications along the [unreadable] globin locus: a) human fetal erythroid x MEL cell hybrids which display a ? to [unreadable] switch with a time course similar to that in vivo will be used to identify (with high throughput technologies determining DNAse I sensitivity, DNA methylation and chromatin histone modifications) the location of the postulated developmental clock of switching. b) [unreadable]YAC x MEL hybrids will be used to experimentally verify the location of the developmental clock. c) Human mutations affecting sequences of the postulated clock will be used to test effects on switching in the YACxMEL or lymphoid x GM-979 hybrid cell models. In SA2 we will determine the in vivo function of regulatory elements of the LCR with studies of [unreadable] thalassemias not due to mutations of [unreadable] gene sequences. 106 cases already available will be searched for point mutations or deletions of core elements of the HSs. Lymphoid x MEL hybrids will determine whether these mutations are in cis to the [unreadable] gene and are associated with [unreadable] mRNA deficiency. YAC/MEL hybrids will be used to verify the thalassemia effect and to investigate mechanisms. We will also test the hypothesis that the phenotype of certain d[unreadable] thalassemias is due to LCR mutations causing unstable interactions between this regulatory element and the downstream ? globin genes. In SA3 we will develop a new approach for the ex vivo investigation of the molecular control of hemoglobin switching which: 1) will be based on fusion of human ES cells with MEL cells, 2) will allow the longitudinal analysis of e to ? and ? to [unreadable] switch in a time frame similar to that of the in vivo human hemoglobin switching and 3) will greatly enhance the molecular analysis of switching through introduction of desired mutations of regulatory elements in human ES cells and testing their effects on switching of human ES cells x MEL cell hybrids.